Air Force Materiel Command

Size: px

Start display at page:

Download "Air Force Materiel Command"

Nathan Lloyd
6 years ago
Views:

1 Air Force Materiel Command ASIP PANEL SESSION Addressing the NDI Crack Miss Problem in Safety of Flight Structures Session Chair John Brausch System Support NDI Lead Materials and Manufacturing Directorate Air Force Research Laboratory I n t e g r i t y - S e r v i c e - E x c e l l e n c e

2 Panel Members Lt. Col. Larry Butkus Deputy Manager USAF Aircraft Structural Integrity Program Aeronautical Systems Center, Wright Patterson AFB Mr. Karl Kraft Propulsion NDI Manager 76th Propulsion Maintenance Wing Oklahoma City Air Logistics Center, Tinker AFB

3 Panel Members Capt. John Schmidt Head, Operational Psychology Department, Naval Operational Medicine Institute Pensacola NAS Mr. John Goglia Director Center for Integrated Emergency Management Saint Louis University

4 The Catalyst In Spring of 2005, a flight restricted Air Force transport aircraft experienced moderate turbulence, which violated the restrictions. Aircraft wing structure was re-inspected per engineering direction. Two cracks found at only 24.3 hours after a previous depot inspection. Reinspections of several other aircraft identified cracks that should have been detected by a previous inspection. Analysis concluded that these cracks were missed during the TCTO inspection. Not isolated to a single base or depot!

5 Cracks Missed Assumed Detection Capability (a NDI ) 0.3 inch Cracks missed >1 inch

6 Challenging Inspection Human Factors (Overhead Inspection) Technical Data (Complex) Surface Condition (Coating Removal Required) Restricted Access (Inspection Area Near Fairing) Large Area (Many Details)

7 NDI Tiger Team Survey NDI Tiger Team Findings AF Wide Issue 4 of 43 AF aircraft systems reporting that nondestructive inspections missed cracks at various locations in past ten years. 42 individual documented misses in past 2 years Misses attributed to multiple factors: Difficult locations to inspect Incomplete inspection procedures Adequacy of oversight Adequacy of training Lack of inspector sensitivity to criticality of task Human Factors Newness of organization to aircraft requirements

8 Air Force Audit Agency Findings 1999 Air Force Audit Agency Report Identified a series of unperformed Inspections 10 bases evaluated 7 bases did not properly accomplish 136 (16%) of 839 inspections inspections 81 (10%) inspections at wrong interval 55 (6%) inspections skipped altogether

9 What Are The Root-Causes? Root-Cause Analysis not routinely accomplished unless mishap occurs Fortunately only one Mishap (Class B) in past ten years related to NDI misses in safety of flight structures Effective Corrective Actions not implemented Have We Been Lucky??

10 What Makes For an Effective Inspection? Well Trained People Empowered People Motivated People Well Engineered Inspections Clearly Defined Requirements Suitable Equipment (Instruments, probes standards) Human Factors Considered in Inspection Development Clear Guidance and Documentation Capability Meets Requirement Strong Organization Employee Feedback Strong Proactive Management Effective Oversight

11 Questions to Address How do we accurately identify the Root Causes and implement effective corrective action? What are the Human Factors driving NDI misses? Are the equipment, procedures and calibration methods effective? Are current training programs effective? Are the current Certification/Qualification programs effective? Is the organization effective? What can be learned from other services or agencies?

12 Undetected Cracks: Structural Significance and Root Cause Investigations ASIP Conference - Panel Session: Addressing the NDI Crack-Miss Problem for Safety-of of-flight Structure 30 November 2005 LtCol Larry Butkus ASC/ENF larry.butkus@wpafb.af.mil

13 Traditional Damage Tolerance Inspection Philosophy Dominant Air Power: Design For Tomorrow Deliver Today Inspections are performed at ½ the time it takes for a crack to grow from some initial (assumed) size [a 0 ] to a critical size [a[ crit ] (i.e. failure) a crit Critical Crack Size Crack Size Inspection Time T i = T f /2 a 0 Initial Crack Size T 0 Time T i Inspection Time T f Times to Failure (when a reaches a crit ) 2

14 Traditional Damage Tolerance Inspection Philosophy Dominant Air Power: Design For Tomorrow Deliver Today If a crack exceeds the detectable crack size [a[ NDI ] and is found and remedial action is taken, then another inspection time is computed Crack Size a crit a NDI a 0 ASIP Damage Tolerance Inspection Intervals Half the time required for a crack to grow to a crit T i1 = 0.5*(T f1 -T 0 ) & T i2 = 0.5*(T f2 -T i1 ) KEY Pre-inspection crack size Remedial action or assumption that a <a NDI Post-inspection crack size (if no remedial action) But what if a crack is missed? Critical Crack Size Detectable Crack Size Initial Crack Size T 0 Time T i1 T i2 Inspection Times T f1 T f2 Times to Failure (when a reaches a crit ) 3

15 Classifying Undetected Cracks Dominant Air Power: Design For Tomorrow Deliver Today The significance of an NDI Miss depends on the lengths of an undetected crack [a miss ] and of a crack that can grow to failure before the next inspection [a[ cr-miss ] ASIP Damage Tolerance Inspection Intervals Half the time required for a crack to grow to a crit T i1 = 0.5*(T f1 -T 0 ) & T i2 = 0.5*(T f2 -T i1 ) Crack Size a crit a cr-miss KEY Pre-inspection crack size Undetected crack size (if a miss = a cr-miss ) Post-inspection crack size (if no remedial action) Type I a miss > a cr-miss Type II a NDI < a miss < a cr-miss Incident Type a NDI a 0 Type III a miss < a NDI T 0 Time T i1 T i2 Inspection Times T f2 Time to Failure (when a reaches a crit ) 4

16 Classifying Undetected Cracks Dominant Air Power: Design For Tomorrow Deliver Today Type I Incident (a (a miss miss > a cr-miss Poses a high risk to flight safety or mission capability Consequences: Failure or potential failure of a safety-of of-flight flight structural component before the next scheduled inspection loss of life or aircraft effects that could result in a Class A mishap Type II Incident miss ) (a NDI < a miss < a cr-miss ) Poses a moderate risk to flight safety or mission capability Consequences: major readiness or economic impacts Type III Incident (a (a miss miss < a NDI Poses a low risk to flight safety or mission capability Consequences: standard repairs to affected structure NDI ) 5

17 Root Cause Investigation Methods Dominant Air Power: Design For Tomorrow Deliver Today Type I Incident (a (a miss miss > a cr-miss Sequential Event and Causal Factor Analysis Walk-Through analysis with Time-Sequence diagram Cause and Effect Analysis Fishbone Diagram Human Performance Evaluation Evaluation of inspectors and management Change Analysis Comparison of expectations with actual events Type II Incident miss ) (a NDI < a miss < a cr-miss Sequential Event and Causal Factor Analysis Cause and Effect Analysis Human Performance Evaluation (optional) Change Analysis Type III Incident (a (a miss miss ) miss < a NDI Informal data gathering & analysis and reporting NDI ) 6

18 Final Thoughts Dominant Air Power: Design For Tomorrow Deliver Today The significance of an undetected crack depends on: The length of the undetected crack The length of a crack that can grow to failure before the next planned inspection The planned inspection interval The consequences of a failure resulting from the undetected crack Various root cause analysis techniques can be used to investigate NDI Miss incidents Guidance has been drafted to assist in root cause analyses Next steps Quantification of a NDI Linking to risk assessments 7

19 Classifying Undetected Cracks Dominant Air Power: Design For Tomorrow Deliver Today Incident Type ASIP Damage Tolerance Inspection Intervals Half the time required for a crack to grow to a crit T i1 = 0.5*(T f1 -T 0 ) & T i2 = 0.5*(T f2 -T i1 ) Crack Size a crit KEY Pre-inspection crack size Remedial action taken Undetected crack size (if a miss = a cr-miss ) Post-inspection crack size (if no remedial action) Type I a miss > a cr-miss a cr-miss Type II a NDI < a miss < a cr-miss a NDI a 0 Type III a miss < a NDI T 0 Time T i1 T i2 Inspection Times T f1 T f2 Times to Failure (when a reaches a crit ) 8

Kraft 76 TH PMXG/MAEPE DSN 336-4116 Karl.Kraft@Tinker.

20 OKLAHOMA CITY AIR LOGISTICS CENTER TEAM TINKER NDI COVERAGE: EDDY CURRENT & ULTRASOUND 9 NOV 05 Karl E. Kraft 76 TH PMXG/MAEPE DSN Karl.Kraft@Tinker.AF.Mil I n t e g r i t y - S e r v i c e - E x c e l l e n c e

21 OVERVIEW Overview Manual Flaw Size Coverage Control Eddy Current (ET) and Ultrasonic (UT) Aircraft and Propulsion Depot and Field Summary Aircraft and Propulsion Field Level Only 2

22 Manual Inspection Overview Manual Flaw Size Coverage Control Summary Propulsion components have much lower fatigue crack tolerance Propulsion handheld inspections are typically unplanned implementations Propulsion manual inspections often used to mitigate risk between scheduled maintenance intervals Propulsion part design change often follows inspection development to eliminate risk mitigating NDI procedures Propulsion component design is complex and typically requires one-of-a-kind probe/standard design no off the shelf probes Probe design is restricted by part geometry/access Propulsion probes purchased per acceptance test plan and often require matching to standard to ensure uniformity in inspection sensitivity and repeatability Sensitivity requirements mandate limiting operator control as much as possible to minimize false calls and ensure coverage Inspectors may require special training including successful completion of POD tests Methods like eddy current require the probe pass over flaw for detection ensuring coverage is critical Propulsion probes are designed to cover a specific coverage zone often smaller than the probe diameter Aircraft inspectors are tasked with covering zones many times larger than probe requiring careful attention to scanning and a methodical approach to ensure coverage 3

23 Aircraft Manual ET Surface Scan Overview Manual CRACK SIZE: INCH Flaw Size Coverage Control Summary TWO AXIS SCAN 4

24 Engines Manual ET Scanning Overview Manual Flaw Size Coverage Control Summary ~CRACK SIZE: >60 MILS SINGLE AXIS SCAN 5

25 Manual UT Scanning Overview Manual Flaw Size Coverage Control Summary ~CRACK SIZE: >30 MILS SINGLE AXIS & GO/NO-GO SCANS 6

26 Flaw Size Overview Manual Flaw Size Coverage Control ASIP largely uses assumptions ENSIP based more upon POD Summary ENSIP flaw sizes roughly 10% of ASIPs 7

27 Ensuring Coverage Overview Manual Flaw Size Coverage Control Summary Scan control increases at depot compared to field Parts disassembled from aircraft or engine Aircraft UT subsurface (fatigue, corrosion, bonding, etc) Propulsion ET fatigue cracks Automation increases at depot to ensure coverage Field inspections Aircraft scan control Less mechanical control Inspections more operator dependent as inspector more often has two axis of freedom to move probe Propulsion scan control More mechanical control Access limitations require special tooling to manipulate probe Reduced operator dependence as inspector is restricted to single axis or no axis of freedom to move probe 8

28 Automated ET Overview Manual Flaw Size Coverage Control Summary CRACK SIZE: ~5-30 MILS BORE AND HOLE SCANS 9

29 Automated UT Scanning Overview Manual Flaw Size Coverage Control Summary MULTI-AXIS SCAN 10

30 Aircraft Semi-Automated UTScan Overview Manual Flaw Size Coverage Control Summary Disbonds & Water Entrapment 11

31 Scanning Categories Overview Manual Flaw Size Coverage Control Summary 2 Axis Manual freehand no restrictions 1 Axis Manual free to move probe in one direction 0 Axis Manual go-no-go Bolthole Scanning scanner controls rotational speed while operator indexes down hole Controlled Fixture rigid test stand where operator centers probe and controls scan start/stop Semi-Automated operator starts/stops and calibrates system, but data acquisition and motion are controlled by machine Automated operator starts system and system runs to completion for operator buyoff 12

32 Scan Control Comparison Overview ET 2 Axis 1 Axis Go-No-Go Bolthole Scanner Controlled Fixture Semi- Automatic Automatic Manual Flaw Size Aircraft Coverage Engines Control Summary UT 2 Axis 1 Axis Go-No-Go Semi- Automatic Automatic Aircraft Engines Propulsion Data for F , F , F101, F108, F /129/400, F118, TF33 Aircraft Data for B-1B, B-52, E-3A, and C

33 Field Level Comparisons Overview Manual Flaw Size ET Aircraft 2 Axis 9 1 Axis 4 Go-No-Go 0 Bolthole Scanner 4 Coverage Engines Control Summary UT 2 Axis 1 Axis Go-No-Go Aircraft Engines Propulsion Data for F , F , F101, F108, F /129/400, F118, TF33 Aircraft Data for B-1B, B-52, E-3A, and C

34 Summary Overview Manual Flaw Size Coverage Control Summary Flaw size limits for engines are more stringent ENSIP utilizes measured reliability Scan control is key to ensuring inspection coverage, sensitivity, repeatability, and reliability Automation is employed at depot for aircraft and propulsion while field inspections are manual Manual propulsion inspections more frequently employ measures to control scanning Propulsion manual coverage zones are typically smaller than the probe sensing element Aircraft inspections require organization and attentiveness due to operator dependence and challenge covering large areas with a small diameter probe 15

35 TEAM TINKER

36 Human Factors Issues in Aircraft Inspection CAPT John K. Schmidt, PhD MSC USN Naval Aerospace Medical Institute DET

38 Naval Aviation Class A Flight Mishaps FY50-04 Class A Flight Mishaps /100,000 Flight Hours 776 Aircraft Destroyed in Angled Carrier Decks Naval Safety Center Naval Aviation Maintenance Program Engineering & Administrative Controls 65 Class A Flight Mishap rate has range between 2-3 mishaps per 100,00 flight hours for the last decade 80% of Naval Aviation Class A Flight Mishaps Involve Human Factors FRS Concept Initiated NATOPS Program Squadron Safety Program System Safety Aircraft 27 Aircraft Destroyed & 19 Deaths in

39 Naval Aviation Human Factors Quality Management Board Human Error Analyses Safety Climate Assessment Best Practices Benchmarking Recommendations for Intervention Initial emphasis on aircrew, but focus quickly expanded to include maintenance

40 FAA Human Factors in Aviation Maintenance and Inspection website provides access to research products, training materials, etc.

41 Aloha Airlines APR 88

42 Watershed Event

43 ttp:// ATA provides free access to ATA SPEC 113: aintenance Human Factors Program Guidelines which outlines the ssential elements for combating maintainer error: Error Investigation & Intervention Ergonomic Audits & Task Redesign Tailored Human Factors Training

45 HFAMI Inspection Reports* The Maintenance Technician in Inspection Human Reliability in Aircraft Inspection Correlates of Individual Differences in NDI Performance Training for Visual Inspection of Aircraft Structures Evaluating the Visual Environment in Inspection Design of the Aircraft Inspection Information Environment HF Good Practices in Borescope Inspection HF Good Practices in Fluorescent Penetrant Inspection *Primarily the work products of Dr. Colin Drury SUNY Buffalo

46 Step Generic Inspection Task Descriptions Initiate* Access Search** Decision*** Respond Repair Reinspect Visual Example Get and read job card & understand covered Area Locate area on A/C; get into correct position Move eyes across systematically, stop if any indication Compare indication against standard in memory (e.g., corrosion) Mark defect and write-up repair sheet, if no defect return to search Drill out and replace rivet Visually inspect marked areas (Adapted from Drury The Maintenance Technician in Inspection) NDI Example Get and read job card, understand covered area, & calibrate Locate area on A/C; get self and equipment into correct position Move probe over each rivet head, stop if any indication Reprobe while closely watching eddy current trace Mark defect and write-up repair sheet, if no defect return to search Drill out rivet, inspect rivet hole, drill out for oversized rivet Visually inspect marked areas

47 Potential Strategies for Improving Inspection Step Changing Inspector Changing System Initiate* Access Search** Decision*** Respond Training in NDI calibration (Procedures) Training in area location (Knowledge & recognition) Training in visual search (Cueing & spatial orientation) Decision training (Feedback & standards) Train writing skills -Redesign job/work cards -Calibration of NDI equipment -Feed forward of expected flaws -Better support stands -Better area location system -Better Location of NDI equipment -Improved lighting -Optical aids -Improved NDI templates -Maintain Inspection Standards -Pattern recognition job aids -Improved feedback to inspector -Improved fault marking -Hands free fault recording (Adapted from Drury The Maintenance Technician in Inspection)

48 Human Factors in Aircraft Inspection Example Considerations: Ergonomic Lighting, Access, Obstructions, Etc. Training Initial (Formal vs. OJT), Certification, & Recurrent Supervision Engineering support, Span of control, QA interaction, etc. Reliability Stress, Fatigue, IDs, Expectations, & Speed/Accuracy Trade-off Work Aids PUBs, Work Cards, Equipment, Etc.

49 Potential for Defense Breaches Ideal Defenses Perceived Process HAZARDS Actual Process Reduced Safety Margins Accepted Minor Deviations Absent/ Failed Defenses Incident Potential

Unsafe Organizational Influences Unsafe Supervision Latent Conditions -Poor Job/Work Card Design -Trained Personnel shortage Latent Conditions Unsafe Conditions -Inadequate Inspection Training -Over

50 Unsafe Organizational Influences Unsafe Supervision Latent Conditions -Poor Job/Work Card Design -Trained Personnel shortage Latent Conditions Unsafe Conditions -Inadequate Inspection Training -Over extended Span of control Latent Conditions -Artisan Inadequately Trained -Artisan Physically Fatigued Failed or Absent Defenses Unsafe Acts Active Failures -Artisan Using Wrong Technique -Artisan Misses Fatigue Crack Reason s Swiss Cheese Model Mishap

FAA/HSAC PART 135 SYSTEM SAFETY RISK MANAGEMENT SAFETY ELEMENT REQUIRED INSPECTION ITEMS (RII) JOB AID

FAA/HSAC PART 135 SYSTEM SAFETY RISK MANAGEMENT SAFETY ELEMENT REQUIRED INSPECTION ITEMS (RII) JOB AID SAFETY ELEMENT 1.3.4 REQUIRED INSPECTION ITEMS (RII) JOB AID The Federal Aviation Administration (FAA) is proactively moving away from compliance based safety surveillance programs to Systems Safety Risk